Feature compensation based on switching linear dynamic model
Nam Soo Kim, Woohyung Lim, Richard M. Stern · IEEE Signal Processing Letters · 2005
In this letter, we propose a novel approach to feature compensation for robust speech recognition in noisy environments. We employ the switching linear dynamic model (SLDM) as a parametric model for the clean speech distribution, which enables us to exploit temporal correlations inherent in speech signals. Both the background noise and clean speech components are simultaneously estimated by means of the interacting multiple model (IMM) algorithm.